iam-git / WellMet (public) (License: MIT) (since 2021-08-31) (hash sha1)
WellMet is pure Python framework for spatial structural reliability analysis. Or, more specifically, for "failure probability estimation and detection of failure surfaces by adaptive sequential decomposition of the design domain".
Clone URLs: https://rocketgit.com/user/iam-git/WellMet ssh://rocketgit@ssh.rocketgit.com/user/iam-git/WellMet git://git.rocketgit.com/user/iam-git/WellMet
release trunk
List of commits:
Subject Hash Author Date (UTC)
convex_hull: implement Grick convex hull aproximation d09ca2e6aca41cfc67f3303eb97b97dd9c6f1fba I am 2022-10-10 03:15:52
qt_gui.qt_gui: distance matrix viewer is ready a532e5259c081a1aff193d3b61e0104c42b08f20 I am 2022-10-08 20:52:14
qt_gui.qt_gui: prepare distance matrix window 4eaea3305643c60c35f0b22421a7e57dbe7a7817 I am 2022-10-08 02:59:01
qt_gui.qt_graph_widgets: set up x limits for all graph widgets 413d114f6a4e4b45c6cd44959cc42588bb7c8008 I am 2022-10-07 22:54:46
qt_gui.qt_plot: set up equal aspect by default 11bfe09b3e96881ef6f7c8202a6723b6e8adc03a I am 2022-10-07 21:41:18
qt_gui.qt_gui: hide "connect/disconnect" button. It wasn't really supported d9f195fe0611d6cf9a8154f0175c6c9506d56247 I am 2022-10-07 21:33:07
qt_gui.qt_plot: fix zero r cirkle (off by one error) 681397de6d075b693897d5f64004c0eb481fd36f I am 2022-10-07 20:34:44
qt_gui: replace globally QtGui by QtWidgets 969a659ae31728e75f587f25f54bab1ae5d6bfff I am 2022-10-07 12:45:34
qt_gui: add radia in Gaussian space (GRaph) widget e3f5be57dcdc6bdd06cda0594ffd795ef5bc8ccc I am 2022-10-07 01:15:38
testcases.testcases_2d: fix natafm_plane pf 75006ccd448c9717c13659e8ceb68800fbda39a7 I am 2022-06-26 12:03:39
misc: add entropy formula 72736d485b4f0fa912a47a49f72954e1ec06c03f I am 2022-06-26 12:02:52
testcases.gaussian_2D: one more product... e835ab15321169c2da085a6bb74e1ed749e67f14 I am 2022-05-11 15:38:24
whitebox: add explicitly solved Gaussian 2D product distribution 23d243baed0b7b71ba887444665420d673dc3fe9 I am 2022-05-11 15:37:19
whitebox: simplify Bessel solutions d097f681f9c96d7f8c6e0c044479681bae110e98 I am 2022-05-08 03:33:44
reorganize the repo 5226fda5d5cba79b6d3471540a4a5f47109d0061 I am 2022-04-30 17:11:21
remore user stuff from the repo ffd15c3d866496bccc5358797d71f2783118235c I am 2022-04-30 16:49:03
dicebox.Goal.get_pf_estimation: fix global stats 423b68cea2d15fce8cf4a157de0160bf26987088 I am 2022-04-21 12:38:48
ghull.Shell_1DS: use geomspace placing if r is large enough. Asi funguje ac4735a382524b05a76e258d774b0e5c0bfd0ec4 I am 2022-04-21 12:37:04
testcases.testcases_2D: add beta_circle_2D case f59387e9981217fde69ba4d87a05ca4017fb5159 I am 2022-04-21 11:40:17
mplot.mart: make isodensities more reliable e18d78f623ef5893a9ecf9cb7e6309de7df7352c I am 2022-04-13 06:18:50
Commit d09ca2e6aca41cfc67f3303eb97b97dd9c6f1fba - convex_hull: implement Grick convex hull aproximation
Author: I am
Author date (UTC): 2022-10-10 03:15
Committer name: I am
Committer date (UTC): 2022-10-10 03:15
Parent(s): a532e5259c081a1aff193d3b61e0104c42b08f20
Signer:
Signing key:
Signing status: N
Tree: a2bac60849500f521d61a9df4e7a2eaa91cebce8
File Lines added Lines deleted
wellmet/convex_hull.py 235 1
File wellmet/convex_hull.py changed (mode: 100644) (index 066a2d1..fae1de0)
... ... def _get_nD_orth_basis(hull):
173 173 #č bůhví, jestli obálka negeneruje A a b-čko dynamicky #č bůhví, jestli obálka negeneruje A a b-čko dynamicky
174 174 b = hull.b b = hull.b
175 175 A = hull.A A = hull.A
176 ndim = hull.sample.nvar
176
177 return _get_nD_orth_basis_from_normals(A, b)
178
179
180 def _get_nD_orth_basis_from_normals(A, b):
181 _nplates, ndim = A.shape
177 182
178 183 #č pro 2D "nastav první a pak po veškerem zbytku #č pro 2D "nastav první a pak po veškerem zbytku
179 184 #č ještě ten předposlední vektor" špatně funguje #č ještě ten předposlední vektor" špatně funguje
 
... ... def get_2FORM_equations(hull):
313 318 b = np.array([[x_backward],[-x_forward]]) b = np.array([[x_backward],[-x_forward]])
314 319 return np.hstack((A,b)) return np.hstack((A,b))
315 320
321 def get_bp_bm(A, points):
322 B = A @ points.T
323 bp = np.nanmax(B, axis=1)
324 bm = np.nanmin(B, axis=1)
325 return bp, bm
326
316 327
317 328 #č jistě musíme mít nějaký zbytečný kus kódu #č jistě musíme mít nějaký zbytečný kus kódu
318 329 #č třida jen pro formu, jen tak na hračku #č třida jen pro formu, jen tak na hračku
 
... ... class QHull:
918 929 return 1 return 1
919 930
920 931
932 # Gaussian brick
933 #č Gaussovská cihla
934 #
935 # it was a mistake to design previous classes without explicit update()
936 class Grick:
937 # take some global functions
938 fire = fire
939 shot = shot
940
941
942 def get_FORM_outside(self):
943 return stats.norm.sf(self.get_r())
944
945
946 def __init__(hull, sample, direct_plan, nrandom=50, auto_update=True):
947 hull.sample = sample
948 hull.direct_plan = direct_plan
949 hull.space = 'G'
950 hull.ndim = sample.nvar
951 hull.nrandom = nrandom #č počet náhodných normalových věktorů
952
953 hull.auto_update = auto_update
954
955 hull.regen()
956
957
958 def regen(hull):
959 hull._npoints = 0
960 hull.mins = np.full(hull.ndim, np.inf)
961 hull.maxs = np.full(hull.ndim, -np.inf)
962
963 hull.bp_dp = np.full(len(hull.direct_plan), -np.inf)
964 hull.bm_dp = np.full(len(hull.direct_plan), np.inf)
965
966 hull.bp = np.full(hull.ndim, -np.inf)
967 hull.bm = np.full(hull.ndim, np.inf)
968
969 hull.orth_basis = np.eye(hull.ndim)
970
971 hull.update()
972
973
974 def _update(hull):
975 if hull.auto_update:
976 hull.update()
977
978 # it was a mistake to design previous classes without explicit update()
979 def update(hull):
980 if hull._npoints < hull.npoints:
981 hull._points = hull.sample.G
982
983 new_points = hull._points[hull._npoints:]
984
985 new_mins = np.nanmin(new_points, axis=0)
986 new_maxs = np.nanmax(new_points, axis=0)
987
988 np.nanmin((new_mins, hull.mins), axis=0, out=hull.mins)
989 np.nanmax((new_maxs, hull.maxs), axis=0, out=hull.maxs)
990
991 new_bp, new_bm = get_bp_bm(hull.direct_plan, new_points)
992 np.nanmax((new_bp, hull.bp_dp), axis=0, out=hull.bp_dp)
993 np.nanmin((new_bm, hull.bm_dp), axis=0, out=hull.bm_dp)
994
995
996 new_bp, new_bm = get_bp_bm(hull.orth_basis, new_points)
997 np.nanmax((new_bp, hull.bp), axis=0, out=hull.bp)
998 np.nanmin((new_bm, hull.bm), axis=0, out=hull.bm)
999
1000 hull._gen_orth_basis()
1001
1002 hull._npoints = hull.npoints
1003
1004
1005 def gen_random_Ab(hull):
1006 ns = hull.ndim * 2 + len(hull.direct_plan) + hull.nrandom
1007 #č necháme vygenerovat maticu náhodných čisel
1008 #č ale použijeme jen část z ní
1009 #č moje zkušenost je, že np.ones a np.random.randn běží stejně rychle
1010 A = np.random.randn(ns, hull.ndim) #random directions
1011 A[:hull.ndim] = np.eye(hull.ndim)
1012 A[hull.ndim:hull.ndim*2] = hull.orth_basis
1013 A[hull.ndim*2:ns-hull.nrandom] = hull.direct_plan
1014
1015 # rand_dir: prepare ns random directions on a unit d-sphere
1016 rand_dir = A[ns-hull.nrandom:]
1017
1018 lengths = np.sum(np.square(rand_dir), axis=1)
1019 lengths = np.sqrt(lengths, out=lengths) #lengths of each radius-vector
1020
1021 # scale all radii-vectors to unit length
1022 # use [:,None] to get an transposed 2D array
1023 np.divide(rand_dir, lengths[:,None], out=rand_dir)
1024
1025 b = np.empty(ns)
1026 #č Nám jde o basis. Dáme ty nejlepší b-čka.
1027 #č znaménka vektorů, tj. směry dopředu nebo dozadu v této fázi nevadí
1028 np.nanmax((-hull.maxs, hull.mins), axis=0, out=b[:hull.ndim])
1029 np.nanmax((-hull.bp, hull.bm), axis=0, out=b[hull.ndim:hull.ndim*2])
1030 np.nanmax((-hull.bp_dp, hull.bm_dp), axis=0, out=b[hull.ndim*2:ns-hull.nrandom])
1031
1032 rbp, rbm = get_bp_bm(rand_dir, hull._points)
1033 np.nanmax((-rbp, rbm), axis=0, out=b[ns-hull.nrandom:])
1034
1035 return A, b
1036
1037
1038 def _gen_orth_basis(hull):
1039 A, b = hull.gen_random_Ab()
1040
1041 #č je to špatné z hlediska testování,
1042 #č ale nevím jak z toho.
1043 #č pro 2D musím udělat vyjimku z "předposlední"
1044 if hull.ndim > 2:
1045 hull.orth_basis = _get_nD_orth_basis_from_normals(A, b)
1046 else:
1047 to_fire = np.nanargmax(b)
1048 a = A[to_fire]
1049 x, y = a
1050 hull.orth_basis = np.array([[x, y],[-y, x]])
1051
1052
1053 #č musí tam bejt G coordinates
1054 hull.bp, hull.bm = get_bp_bm(hull.orth_basis, hull._points)
1055
1056
1057
1058
1059
1060
1061
1062
1063 def get_orth_outside(hull):
1064 hull._update()
1065 return calculate_brick_complement_probability(hull.bm, hull.bp)
1066
1067
1068 def get_orth_equations(hull):
1069 hull._update()
1070 return hull.equations
1071
1072
1073 def get_2FORM_outside(hull):
1074 hull._update()
1075 return stats.norm.cdf(hull.bm[0]) + stats.norm.sf(hull.bp[0])
1076
1077 def get_2FORM_equations(hull):
1078 hull._update()
1079 A = np.vstack((-hull.orth_basis[0], hull.orth_basis[0]))
1080 b = np.array([[hull.bm[0]],[-hull.bp[0]]])
1081 return np.hstack((A,b))
1082
1083
1084 @property
1085 def points(hull):
1086 if hull.auto_update:
1087 return getattr(hull.sample, hull.space)
1088 else:
1089 return hull._points
1090
1091
1092 @property
1093 def npoints(hull):
1094 if hull.auto_update:
1095 return len(hull.sample)
1096 else:
1097 return hull._npoints
1098
1099
1100
1101 @property #č stejně nikdo to nebude spouštět
1102 def nsimplex(hull):
1103 return hull.sample.nvar * 2
1104
1105
1106 @property
1107 def A(hull):
1108 hull._update()
1109 #č žádná optimizace, ale stejně nikdo to nebude spouštět
1110 return np.vstack((-hull.orth_basis, hull.orth_basis))
1111
1112 @property
1113 def b(hull):
1114 hull._update()
1115 return np.concatenate((hull.bm, -hull.bp))
1116
1117 @property
1118 def equations(hull):
1119 hull._update()
1120 #č žádná optimizace, ale stejně nikdo to nebude spouštět
1121 A = np.vstack((-hull.orth_basis, hull.orth_basis))
1122 b = np.concatenate((hull.bm, -hull.bp))
1123 return np.hstack((A,b[:,None]))
1124
1125 def is_inside(hull, nodes):
1126 hull._update()
1127 x = nodes.G
1128
1129 #E [normal, offset] forming the hyperplane equation of the facet (see Qhull documentation for more)
1130 bp = np.atleast_2d(hull.bp).T
1131 bm = np.atleast_2d(hull.bm).T
1132
1133 # N=ns, E - number of hyperplane equations
1134 ExN = hull.orth_basis @ x.T
1135 higher = np.all(ExN < bp, axis=0)
1136 lower = np.all(ExN > bm, axis=0)
1137 return np.all((higher, lower), axis=0)
1138
1139 def is_outside(hull, nodes):
1140 return ~hull.is_inside(nodes)
1141
1142 def get_design_points(hull):
1143 hull._update()
1144 sample_model = -hull.A * hull.b.reshape(-1,1)
1145 return hull.sample.f_model.new_sample(sample_model, space=hull.space)
1146
1147 def get_r(hull):
1148 hull._update()
1149 "calculates minimal signed distance to planes. Can therefore be negative"
1150 return min(-hull.bm[0], hull.bp[0])
1151
1152
1153
1154
921 1155
Hints:
Before first commit, do not forget to setup your git environment:
git config --global user.name "your_name_here"
git config --global user.email "your@email_here"

Clone this repository using HTTP(S):
git clone https://rocketgit.com/user/iam-git/WellMet

Clone this repository using ssh (do not forget to upload a key first):
git clone ssh://rocketgit@ssh.rocketgit.com/user/iam-git/WellMet

Clone this repository using git:
git clone git://git.rocketgit.com/user/iam-git/WellMet

You are allowed to anonymously push to this repository.
This means that your pushed commits will automatically be transformed into a merge request:
... clone the repository ...
... make some changes and some commits ...
git push origin main